United States Patent [19]

[11] Patent Number:

McMilleu

[45]

[54]

Date of Patent:

4 9 531 a 586
Jul. 30, 1985

METHOD OF SOLVENT STIMULATION OF

3,554,289

1/1971

Webb ............................ .1 166/305 R

HEAVY 01L RESERVOIRS

3,881,550

5/1975

Barry

..... .. 166/303 X

3,978,926 9/1976 Allen

[75]

Inventor:

[73]

Asslgmeiz

James M. McMillen, Arlington, Tex.

4,280,559

7/1981

4,293,035

10/1981

{1:149:11 011 Cmpmatmn, New York

[211 App]. No: 564,044

[22]

Filed:

166/305 R X

Best . . . . . .

. . . . . . . . . ..

Fitch

166/303

166/271 X

4,450,913 5/1984 Allen et a1. ........... .. 166/267 x

FOREIGN PATENT DOCUMENTS

600 21 8 6/1960 Canada

Dec. 21 1983

.............................. ..

166/263

Primary Examiner-George A. Such?eld

Related US, Application Data
[63]

Attorney, Agent, or Firm-Alexander J. McKillop;

Continuation-in-part of Ser. No. 307,544, Oct. 1, 1981,

abandonedInt. GL3 ..................... ..

[58]

invention provides a method for cyclic Solvent

U:S. Cl. ............................... ..

Field of Search .......... .. 166/263, 267, 303, 305 R

[56]

References Cited

stimulation of production of heavy

from an under
ground reservoir penetrated by a well which comprises:

(a) injecting into the reservoir a liquid solvent having

a ratio of crude viscosity to solvent viscosity of at

U.S. PATENT DOCUMENTS

2,258,614 10/1941
2,259,428 10/1941
2,369,831 2/1945

Mlchael G Glman; Stamslaus Aksman

[57]
ABSTRACT

least about 10 and in an amount between about 5

Kendrick .......................... .. 166/263

Shelley ...... ..
166/305 R X
Jones et al. ............... .. 166/305 R

2,927,639

3/ 1960 Schuessler et a1. .

166/305 R

2,998,066

8/1961

Nixon, Sr. ......... ..

166/305 R

3,295,601

1/ 1967

Santourian ........................ .. 166/263

barrels and about 25 barrels per foot of oil-bearing

formation; and
(b) thereafter producing a solvent-oil mixture.

10 Claims, No Drawings

4,531,586

erably 100 or more. Suitable solvents are light crude oil,

syncrude, diesel fuel, condensate, cutter stock, or other

METHOD OF SOLVENT STIMULATION OF

HEAVY OIL RESERVOIRS
CROSS-REFERENCE TO RELATED
APPLICATION

This application in a continuation-in-part of copend

ing application Ser. No. 307,544, ?led Oct. 1, 1981, now
abandoned, the contents of which are hereby expressly

incorporated herein by reference.

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention is concerned with the stimulation of

production of heavy oil.

light hydrocarbons. It is within the contemplation of

5

this invention that about a third of the injected solvent

can be solvent-rich production, i.e., the initial produc
tion from a solvent stimulated production that is rich in
solvent content. The injected solvent having the desired
viscosity moves away from the wellbore into the forma
tion forming solvent ?ngers that are larger near the
wellbore and decrease in size as they advance into the
formation. The solvent ?ngers near the wellbore pro

vide a high mobility path for subsequent produced oil

backflowing into the well. To provide adequate solvent
?ngering near the wellbore and signi?cantly reduce the

viscosity of the in-place oil by diffusion and mixing, the

amount of solvent that is injected is between about 5

barrels and about 25 barrels per foot of oil-bearing for
mation, preferably between about 10 and about 20.
After solvent injection has been completed, there
should be little or no soak time, i.e., the time between
the end of solvent injection and the start of production.
Generally, the soak time will be between about an hour
or less and about 48 hours, preferably less than 24 hours.
In accordance with this invention, there is little advan
were designed to help produce. Reported results have 25 tage, if any, in an appreciable soak time to effect diffu
indicated that well clean-up and emulsion reduction
sive mixing of solvent and heavy oil. It appears that
have been the primary mechanisms responsible for any
prolonged soak time of several days or more isolates

2. Description of the Prior Art

Solvents have been injected into heavy oil wells in
years past in order to improve production from those
wells. Generally, the solvents used were light hydrocar
bons, such as a rich gas, or light liquid hydrocarbons
plus additives. These additives were for such purposes
as dissolving deposited solid hydrocarbons or for break
ing oil-water emulsions. Such mixtures were much
more expensive than equal volumes of the crude they

increased production resulting from solvent injection.

Production improvement as a result of the viscosity

reduction due to solvent dilution has been discounted as

a practical procedure.
It is the unexpected discovery of this invention that
the production of heavy oil can be greatly increased by
solvent dilution of heavy crude alone, the dilution re

sulting in a reduced viscosity of the solvent/crude mix

ture. The production increase is a result of proper de

solvent by gravity driven mixing and destroys the high

mobility paths near the well, so that there is little in
crease in production over that obtained in unstimulated

production.
Production, when commenced after solvent injection
is completed, should be maintained continuously. Any
shut-down should be kept under48 hours, preferably
under 24 hours. Production should be continued until

the fraction of solvent in the produced oil has dropped

to about 12% or less, regardless of production rate. At
this point, additional solvent or solvent and solvent-rich
production
can be injected into the reservoir followed
40
production, once begun, is also important to the proce

sign criteria of the solvent stimulation procedure, in

volving the volume of solvent used and the soak period
prior to resumption of production. The continuity of
dure. It has also been found that the same well can be

by resumed production.

These cycles of solvent injection and production can

be
continued until the reservoir is exhausted. Essentially
on the choice of proper intervals between solvent injec
all (about 97%) of the solvent injected into a formation
tions. In addition, it has been determined that relatively
inexpensive hydrocarbons without additives, such as 45 in a multi-cycle solvent stimulation is returned with
produced oil. Thus, it can be separated from the pro
light crude oils, can be used as the solvent.

solvent stimulated successfully many times, depending

duced oil on site, if desirable or practical, by some sepa

SUMMARY OF THE INVENTION

ration method, e.g., in a topping plant, and used in sub

sequent injections. Alternatively, the mixture of solvent
This invention provides a method for cyclic solvent
stimulation of production of heavy oil from an under 50 and heavy oil can be used directly as a re?nery charge
stock and it has the advantage of being easily pipelined. _
ground reservoir penetrated by a well which comprises:
(a) injecting into the reservoir a liquid solvent having
EXAMPLE
a ratio of crude viscosity to solvent viscosity of at
least about 10 and in an amount between about 5

barrels and about 25 barrels per foot of oil-bearing

formation; and
(b) thereafter producing a solvent-oil mixture.
DESCRIPTION OF SPECIFIC EMBODIMENTS
As used in speci?cation and claims, heavy crude oil
is a viscous crude oil that has poor flow characteristics
in the reservoir. In general, it is a crude oil that has a
API gravity of about 20 degrees or lower.
The solvent used should be substantially, but not

necessarily completely, miscible with the crude oil. It

must have viscosity lower than that of the crude. In
general, the ratio of crude viscosity to solvent viscosity
at reservoir conditions should be at least about 10, pref

The well used was a Cox Penn Sand Unit well lo

cated in Carter County, Okla. It is completed in the 4th

Deese sand which contains oil with a gravity of 15
API. The thickness of the oil bearing formation is about
43 feet. The unstimulated rate of production from this
well was 8.5 barrels of oil per day (BOPD). The solvent
used was Graham-Deese light crude oil (about 34
API), unless otherwise noted.
The well was subjected to ?ve injection- production
cycles. The ?rst three cycles each consisted of solvent
injection of about 400 barrels. Cycle 1 used as solvent
fresh Graham-Deese crude for the full volume. The ?rst

135 barrels of solvent-rich production (greater than 28

API, containing about 70% solvent) was saved and used
as the middle portion of the following injection, the

4,531,586

remainder being the fresh light crude. This procedure

was followed through the subsequent cycles. Each
cycle involved injection one day with production start
ing the next morning for convenience in monitoring the

rels/barrel of solvent injected.

Although the present invention has been described
with preferred embodiments, it is to be understood that

returning ?uids.

modi?cations and variations may be resorted to, with

out departing from the spirit and scope of this invention,

creased average production rate or incremental oil bar

For the fourth cycle, the injected volume was in

as those skilled in the art will readily understand. Such

modi?cations and variations are considered to be within

creased to about 1200 barrels to determine whether

average production rate or the incremental barrel/in

the purview and scope of the appended claims.

What is claimed is:
1. A method for solvent stimulation of heavy crude
oil from an underground reservoir penetrated by a well,

jected barrel ratio was changed using a larger injected

volume. Both were dramatically lower, showing that
the injected volume can be too large and that multiple
small injections produce more incremental oil.
The ?fth cycle, in order to make a direct comparison
with the ?rst three, was again an about 400 barrel injec 15
tion. The initial production rate was too high and aver
age rate and incremental volumes suffered in compari
son, but incremental oil was produced and increased
average rate achieved more nearly like the ?rst three

400 barrel cycles than the 1200 barrel cycle.

These runs and the results are set forth in the follow

such a viscosity that the ratio of the crude oil vis

cosity to solvent viscosity is at least about 10, to
form solvent ?ngers near the well, and in an
amount between about 5 barrels and about 25 bar

rels per foot of oil-bearing formation to provide

adequate solvent ?ngering near the wellbore and

signi?cantly reduce the viscosity of the heavy

crude oil; and
(b) thereafter producing a solvent-oil mixture.

ing Table:
TABLE
Injected
Volume
(Bbl

which comprises:
(a) injecting into the reservoir a liquid solvent having

2. The method of claim 1, wherein said ratio is at least

Total

Incre
mental

Production

Production

Average

Test

of
Solvent

3. The method of claim 2, wherein said amount is

between about 10 barrels and about 20 barrels per foot

of oil-bearing formation.
4. The method of claim 3, wherein the producing in

Fraction

Cycle

Solvent)

(Bbl
Oil)

(Bbl
Oil)

Rate
(BOPD)

Period
(Days)

Recover
ed %

1
2
3
4
5

385
401
394
1181
401

921
935
1022
2302
892

230
227
280
350
185

14.9
14.8
15.0
12.3
13.6

36
36
42
91
36

89
97
97
97
101

pm-stimulation rate was 8.5 BOPD.

25 about 100.

30 step (b) is carried out until the amount of solvent in the

produced solvent-oil mixture drops below about 12

percent.
5. The method of claim 4, wherein steps (a) and (b)
35

are repeated.
6. The method of claim 5, wherein the solvent is a

light crude oil.

From the data in the Table, it will be noted that the

four cycles of about 400 injected barrels of solvent
produced an incremental 185 to 280 barrels of oil (0.48

7. The method of claim 1, wherein at least a portion

of the injected solvent is solvent-rich production from a

previous cycle.

to 0.71 barrels/injected barrel) at an average rate of 5.3 40

8. The method of claim 7 further including the step of
to 6.8 BOPD greater than the unstimulated rate of 8.5
allowing the reservoir to undergo a brief soak time
BOPD. This amounts to a 62% to 80% increase in pro
ranging from about one hour to about 48 hours after

duction. The 1181 barrel injection produced 350 barrels

solvent injection according to step (a).

of incremental oil (0.3 barrels/barrel) at a rate of 3.8

9. The method of claim 8, wherein the soak time is
BOPD greater than the unstimulated rate, a 45% in~ 45 between about one hour and about 24 hours.
crease in production.
10. The method of claim 5 wherein the solvent is light
Additional experiments in seven other wells in ?ve
crude oil, syncrude, diesel fuel, condensate, cutter stock
other heavy oil reservoirs have shown results similar to
or other light hydrocarbons.
*
*
*
*
*
or better than those cited above, in terms of either in
50

55

65